Youtube worked for me fine without desktop effects on.
But after effects turning on - it also started to make problems - playback wasnt visible (I saw only the frame last before maximizing but deformed and Flash's note to press Escape to switch back to window) and screen was flickering.

Created attachment 374378 [details]
Fragments on screen

When I watched video on youtube on fullscreen (for few seconds) and then escaped it and X didn't totally hang, I was able to capture graphical problems on screen. Hope this helps.

Completely the same with modeset=0 kernel parameter.

(In reply to comment #3)
> Completely the same with modeset=0 kernel parameter.

That would mean that it also occurs with fbdev driver (no 2D acceleration at all). Weird.

(In reply to comment #4)
> (In reply to comment #3)
> > Completely the same with modeset=0 kernel parameter.
> 
> That would mean that it also occurs with fbdev driver (no 2D acceleration at
> all). Weird.

I think that the kernel parameter would need to be 'i915.modeset=0" in order to disable KMS.

(In reply to comment #2)
> Created an attachment (id=374378) [details]
> Fragments on screen
> 
> When I watched video on youtube on fullscreen (for few seconds) and then
> escaped it and X didn't totally hang, I was able to capture graphical problems
> on screen. Hope this helps.

I had exactly the same problem on RC2 (x86_64). I still could do things like pause the video without seeing what was happening.

But now I'm on the final version (x86 now, though). Now using fullscreen flash video hangs the notebook completely (tested with this one: http://www.youtube.com/watch?v=wtJUloE6F78). Sound keeps looping, freezed screen and keyboard lights doesn't responds.

I think that flash video uses openGL for rendering, could the problem be related to something that only flash uses?.

I've tested on the same video without KMS (reverts to FBDEV as Stefan wrote) and could reproduce the problem.

(In reply to comment #5)
> I think that the kernel parameter would need to be 'i915.modeset=0" in order to
> disable KMS.

Ah, Mea culpa:)
After reboot with this parameter set, the video fullscreen playback works okay.
But using fbdev driver permanently is not possible, due to slowness and so on.

(In reply to comment #6)
> (In reply to comment #2)
> > Created an attachment (id=374378) [details] [details]
> > Fragments on screen
> > 
> > When I watched video on youtube on fullscreen (for few seconds) and then
> > escaped it and X didn't totally hang, I was able to capture graphical problems
> > on screen. Hope this helps.
> 
> I had exactly the same problem on RC2 (x86_64). I still could do things like
> pause the video without seeing what was happening.
> 
> But now I'm on the final version (x86 now, though). Now using fullscreen flash
> video hangs the notebook completely (tested with this one:
> http://www.youtube.com/watch?v=wtJUloE6F78). Sound keeps looping, freezed
> screen and keyboard lights doesn't responds.
> 
> I think that flash video uses openGL for rendering, could the problem be
> related to something that only flash uses?.
> 
> I've tested on the same video without KMS (reverts to FBDEV as Stefan wrote)
> and could reproduce the problem.

oops, I wanted to write that I *couldn't* reproduce the problem. It works without hanging, though horribly slow.

This problem is similar to Bug 623873.

I have this bug too. I tried play a flash game (using Adobe Flash Player) in fullscreen and the same bug occured.

There's a pair of bugs that seem related to this one on FreeDesktop's Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=26937 and https://bugs.freedesktop.org/show_bug.cgi?id=28573.

The same problem occurs in the final version: https://bugzilla.novell.com/show_bug.cgi?id=623239

The same problem occurs in the final version: https://bugzilla.novell.com/show_bug.cgi?id=623239

Please delete the duplicate comments.

*** Bug 623873 has been marked as a duplicate of this bug. ***

I think I've found a workaround.
1. install xorg-x11-driver-video-intel-legacy package
2. change the driver to 'intellegacy' in file /etc/X11/xorg.conf.d/50-device.conf
3. add the 'nomodeset' kernel parameter to your kernel line in /boot/grub/menu.lst 
After a reboot I can watch youtube videos on fullscreen.

Yes, this workaround works for me.

Just got bitten by this bug on my Samsung NC10 and Gateway MX6923b (both use the Intel chipset). The work-around in comment 16 seems to help.

Yes, but I would personaly prefer working with KMS :-)

I have exactly the same bug on my new Intel Core i5 system with integrated video.
I've tried to build a git version of X.org, Mesa, drivers, with no luck.. (((
Nothing seems to help.
Meanwhile, I found, that on Fedora 13, flash works great on my hardware. It's looks strange, because F13 also uses KMS and new (2.10.0) intel drivers.

//sorry for my English )))

I'm testing here for some hours the workaround. Seems to be fine. Tested also without the 'nosetmode' option and there are also no crashes. However, will follow your advice with the option.

HW: Intel Atom N270 + Intel 945GME (Netbook Medion Akoya E1210)

Nice to read that using older drivers solves the problem.
Is there the possibility to push a downgrade/upgrade of the intel driver via online updates? Seems like the version included on 11.3 is buggy; there is another bug report about the intel driver causing problems which can be solved by downgrading: https://bugzilla.novell.com/show_bug.cgi?id=615649 .

*** Bug 624298 has been marked as a duplicate of this bug. ***

*** Bug 623239 has been marked as a duplicate of this bug. ***

I managed to find a temporary solution for this problem, without using the old driver. But it is sufficiently complex.

1 - I patched the sources of the standard kernel (2.6.34-12-desktop) with drm/i915 patches from official 2.6.34.1 and 2.6.34.2 kernels. I have downloaded them from kernel.org site (via gitweb).
After that i have recompiled the kernel.

2 - I have compiled a git versions of libdrm, xserver, mesa, intel video driver, and all related components. (i've used SuSE Build Service, and I can provide link to repo, if needed)

3 - I have installed a precompiled 32-bit version of Firefox, from mozilla site.

4 - I've noticed that my dmesg said about mtrr problems when starting a i915 driver. I've fixed that too.

Now flash works great in fullscreen. (i tested it with effects turned off).
Perhaps the problem could be solved much easier in your case...

//sorry for my English )))

i've just tested flash with 64-bit version of browser - looks like it works too.

when you tink that the error can be resolved .Sorry for my english
En cuanto tiempo se podria arreglar el error?

(In reply to comment #27)
> when you tink that the error can be resolved .Sorry for my english
> En cuanto tiempo se podria arreglar el error?

I don't know. Most likely the problem will be solved, when newer kernel (with patches from official kernel.org) and maybe newer xorg will be available in official suse repo.

For now, i can build *non official* suse kernel RPM with all patches, and newer xorg (already built), and place it on SuSE Build Service repos. I can't guarantee that this will fix problem with your hardware, and not produce new problems. This bug seems to be complex.
But if you want to try, i'll make all neccesary RPMs.

if kernel fixes from the opensuse-11.3 branch are required, we should do
a kernel update too.

there are quite some fixes already anyway.

I can confirm this is the case for me to.  Very annoying :(

vote++

Could you give just the latest kernel for 11.3 a try first?

http://download.opensuse.org/repositories/Kernel:/openSUSE-11.3/openSUSE_11.3/

I have the same problem. I tried the latest kernel as suggested by Stefan D., but nothing changed.

Jakub?

I've got the same Problem on three diffrent Computers.
All OpenSuse 11.3 KDE 32bit
- eeePC 1000H
- Lenove G550
- Dell Optiplex 280

So, this is no Hardware Porblem.
I'm not very happy with this bug on OpenSuse 11.3.
Please - Fix it...

Your Hardware include Intel's chipsets?

Yes, all these are Intel processors

Arghh! NEEDINFO is set due to comment #31! Still waiting for feedback here.

Sory for the delay, I hope I will finally be able to try new kernel today evening (in aprox. 8 hours)

Hi,

For me, the "new" kernel (2.6.34.4-1.1) seems to work. No freeezing anymore (yet).
Tested fullscreen flash three times over a time of 10 minutes.

Hardware: 
Lenovo Thinkpad T500 
GPU: Mobile Intel GM45 Express Chipset

Installing 2.6.34.4-1.1 would breake dependencies for preload-kmp-desktop. I dont't want to break my notebook. Shall I really test that kernel, is it safe? Thx.

preload-kmp-desktop is not an essential package for your system. You can ignore that dependancy. After testing you can of course reinstall the 2.6.34 RPMs.

Worked with new kernel.

Thanks for verification. So let's close the issue as fixed.

Well, after that, the only way to log into my account is to use safe mode and then from console run "startkde", but I am not sure, whether it is connected with the kernel changes... root and new user I created logged in well :-(

(In reply to comment #44)
> Well, after that, the only way to log into my account is to use safe mode and
> then from console run "startkde", but I am not sure, whether it is connected
> with the kernel changes... root and new user I created logged in well :-(

Some RANDR 1.2 desktop setting configuration which no longer works maybe. I guess it works with the same user in other desktops.

Hhm, flash fullscreen froze today again (first partially but after some seconds keys and mouse didn't work anymore). Anyone else having still problems with the kernel from the repositories?

Again, kernel-2.6.34-12.3 with intel-legacy driver is completely stable.

After changing to kernel-2.6.34.4-6.1 and commenting out 'Driver "inlellegacy"' in /etc/X11/xorg.conf.d/50-device.conf no flash-video is able to crash the kernel. Yes, on my HW(**) it works fine for flash videos, but only for flash!

Because there is still a side effect. After trying to open this internet site(*) in Firefox (3.6.8) the KDE session crashes and suddenly KDM login manager starts up again. This doesn't occur with intel-legacy and it doesn't occur with Opera and Konqueror.

I think, although the flash video problem seems to be solved, the workaround with intel-legacy is still the best solution.

However, does anyone know the bug number where this problem is in discussion?

(*) Example for crash:
http://einestages.spiegel.de/static/topicalbumbackground/3962/tod_am_zaun.html

(**) Just only FYI here my HW description: Medion Akoya E1210 Netbook

lspci
00:00.0 Host bridge: Intel Corporation Mobile 945GME Express Memory Controller Hub (rev 03)
00:02.0 VGA compatible controller: Intel Corporation Mobile 945GME Express Integrated Graphics Controller (rev 03)
00:02.1 Display controller: Intel Corporation Mobile 945GM/GMS/GME, 943/940GML Express Integrated Graphics Controller (rev 03)
00:1b.0 Audio device: Intel Corporation N10/ICH 7 Family High Definition Audio Controller (rev 02)
00:1c.0 PCI bridge: Intel Corporation N10/ICH 7 Family PCI Express Port 1 (rev 02)
00:1c.1 PCI bridge: Intel Corporation N10/ICH 7 Family PCI Express Port 2 (rev 02)
00:1d.0 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #1 (rev 02)
00:1d.1 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #2 (rev 02)
00:1d.2 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #3 (rev 02)
00:1d.3 USB Controller: Intel Corporation N10/ICH 7 Family USB UHCI Controller #4 (rev 02)
00:1d.7 USB Controller: Intel Corporation N10/ICH 7 Family USB2 EHCI Controller (rev 02)
00:1e.0 PCI bridge: Intel Corporation 82801 Mobile PCI Bridge (rev e2)
00:1f.0 ISA bridge: Intel Corporation 82801GBM (ICH7-M) LPC Interface Bridge (rev 02)
00:1f.2 SATA controller: Intel Corporation 82801GBM/GHM (ICH7 Family) SATA AHCI Controller (rev 02)
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8101E/RTL8102E PCI Express Fast Ethernet controller (rev 02)
02:00.0 Network controller: RaLink RT2860
E12

I have this kernel with intel-legacy  and use any ie (konqueror,chrome,etc) and the notebook crash (npviewer freez or kde screen turns inestable or directly de system hangs)sorry my english 

I follow the bug and I see that is a problem of the kernel but Im not sure.

Someone have a temporary fix 

thanks

Opensuse 11.3
Intel 945gm 
Linux 2.6.34-12-desktop x86_64
Kde 4.4.4

(In reply to comment #48)
> I have this kernel with intel-legacy

This bug was about intel, not intel-legacy driver. intel-legacy *sometimes* can serve as workaround for intel driver issues.

Are we likely to see a kernel update in the update repository soon? Presumably the update will also take into account this bug?  Thanks :)

The 11.3 kernel will be updated to 2.6.34.4:
https://build.opensuse.org/maintenance/new_11.3#anker-35403

The update will appear in the test repo soon (http://download.opensuse.org/update/11.3-test/)

Great news, thanks for the quick reply.

Thanks guys for all and sorry for not understand before

Question.You guys or the people that resolve the problems in here are a comunity or employes of novell ? It is simple doubt

Allow me to pile-on.

I've just upgraded by 11.0 install on my ThinkPad x61 to 11.3 from DVD (32 bit).

The full-screen flash lockup (it does *not* seem to affect MPlayer 1.0rc2 fullscreen, frex) is one of several regressions I've seen.

I'm already *using* this kernel:

  2.6.34-12-desktop #1 SMP PREEMPT 2010-06-29 02:39:08 +0200 i686 i686 i386 GNU/Linux

so upgrading to it (comment 51) will clearly not help me much.

My video is:

  00:02.1 Display controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (rev 0c)

so presumably this is a graphics driver bug.

My symptom, too, is a hard hardware-lockup, no response to anything at all except VESA Poweroff; audio continues playing in a loop of what was in the hardware buffer.

Per comment 43, it would be nice if Stefan could enumerate what he thinks is the advised fix for this matter at this time, as it's not possible to extract a clear answer to the question from the comment stream; at least, *I* can't do it.

> 2.6.34-12-desktop #1 SMP PREEMPT 2010-06-29 02:39:08 +0200 i686 i686 i386
> GNU/Linux
>
> so upgrading to it (comment 51) will clearly not help me much.

That's the GA-kernel, *not* the update kernel.

 2.6.34.4 > 2.6.34-12

I should note here, that I was *also* (after my upgrade) seeing screen-flashes-to-black-for-50ms behaviour in Firefox 3.6.8, *mostly* (though not definitively) in AJAXy app pages like Facebook and Twitter; I didn't see it on WX Underground's Full Screen Weather, nor on most to all other pages; I don't believe I saw it on other apps when FF was not focussed.

So the GA kernel is 2.6.34.0-12, and was misnumbered on release.  Got it.

(Note: that's an editorial opinion, based on about 20 years of software release; if you disagree with it, let us not start an R-war, ok? :-)

I will attempt to upgrade my kernel, after figuring out why I haven't already been presented with it by the updater.

And the current expected fix is: "Update your kernel to 2.6.34.4"  Thanks.

Ok.  First, the upgrade to .4 does appear to have fixed the problem, without breaking anything else that I've spotted yet.  Thanks.

That said, 

  zypper up kernel-desktop

wouldn't run because the kernel is in a different repo, and the suggested

  zypper install kernel-desktop-2.6.34.4-6.1.i586

installed *989MB* of kernels, and put *all of them* in my menu.lst, which seems excessive: notably, it did *not* leave the older kernel in the menu.lst, nor indeed in the /boot directory at all.

Does RPM/zypper not have any special handling for kernels which says "you shouldn't uninstall the old one when you install a new one, unlike what's the proper behavior for normal apps"?

One more clarification for people reading the bug later: this problem is not limited to x86-64; it happened on my i686 install as well.

... and I'm reopening.

While my YouTube players and most other Flash video players now seem to be working fine, I did just have the same hang reported by a couple other people up-thread, with the Comedy Central player, on their own site; full episode of The Daily Show.

This was about the 35th or so minute of video I'd watched, across 4 or 5 player loads; the hang was that while the audio continued to the end of the segment, the video froze, and the keyboard was completely ignored (except Fn-F5, which toggled the wireless lights; Fn-F4 suspend and Fn-F12 hibernate were ignored).

Ctrl-Alt-F1 to switch away from X was also ignored.  I infer that the kernel change fixed a *kernel* bug, and that there's now a related userland bug that's at issue.

FWIW: the Comedy Central player worked in an unusual fashion: when going to full-screen, I had to Alt-Tab to get to it, which I've never had to do before.  ESCaping out of it back to embed worked normally.

This is Flash 10.1r82 from repo, in FF3.6.8 from repo, on a 11.3 upgraded from a clean 11.0 install kept upgraded.  I have adjusted the tags on this bug for my particular instance.

In other news: Novell has *the best* looking Bugzilla I have seen; have the CSS patches been released?

Thanks for reopening. That way we'll never be able to get closed any bug. There will always be a person, who thinks to have the same issue with the same culprit
and reopens the bug.

JFYI, I can't see any of these daily shows on the comedy site. It shows me

 "Sorry, the video you are trying to watch is not available in your region."

Only the advertisements I can see.

Stefan: what precisely is your point here?

Did you want me to lose all the context of this bug, which appears to be the same problem, solely on my uninformed personal speculation that it might have a different cause?

The attitude does no one any good, really.  

If you have an *informative comment* to make, like

   "It appears this might actually be a different problem, could you please open a separate bug for it"

then *why don't you say that*, instead of being all pissy and attitudinal about it, eh?

In other news, Comedy Central's regional limitations are really not something I can do anything about.  That said, something's still broken, and I do hope *someone* gives a shit about fixing it; I do my best to write Simon Tatham-quality bug reports, and I do so hate finding out I'm wasting my time.

Jay: One issue one bug report sounds sensible to me. Spamming the bug with comments about zypper not, neither using words like shit or similar attitude.

If one is not 100% certain that it's the same cause, the rule of thumb is to ask whether to open a separate report or not before re-opening a fixed bug report.

You seem a bit sensitive, so I hope the above hint helps you to avoid taking things personal next time you report bugs.

> One issue one bug report sounds sensible to me.

That was my thought, as well; hence my reopening *this* bug (which several other bugs have been marked dupes of), instead of opening yet another new bug about a problem with the same description.

I apologize for the cuss word; clearly zypper had left me irritated. 

That said, I don't think that the comment on its behavior was per se out of scope in this context; the instructions are *install a new version of the kernel*.  I would think that people following this bug would want to be warned that that process might not go as they expect.  Particularly if the result is *downloading a gigabyte of code*; amazingly, there are still people out there on slow DSL links, or even dialup.  *I* would certainly want to be warned.

I don't think my comment 60 shows any oversensitivity at all, actually.  62, perhaps, but the reasons why ought to be intuitively obvious to the most casual observer.

In short, this is the NFL: we (the users) report the bugs, and it is not really within our sphere of responsibility to be familiar with the fine points of style of a specific bugzilla; I have accounts on about nine of them at last count.

I've seen abusive bug comments... mine were not among them.

Meta aside, who is actually responsible for this bug, and do they have enough additional information to continue to work it, or should I provide more?

Further: updating the kernel did *not* fix Youtube's player either; it too will still hang in fullscreen mode, though also it appears to lock X, not the underlying player code or the kernel -- I get audio still coming out, and a working mouse cursor, but the entire keyboard is dead, with the same exceptions mentioned above.

And another data point:

mplayer, which does *not* break in fullscreen mode, complains about this during startup:

Failed to open VDPAU backend libvdpau_nvidia.so: cannot open shared object file: No such file or directory
[vdpau] Error when calling vdp_device_create_x11: 1

Since this laptop has Intel video, I'm not sure why it might be trying; the Xorg and mplayer are both from the manual upgrade (ie: repos); mplayer is

MPlayer dev-SVN-r31930-4.5-openSUSE Linux 11.3 (i686)-Packman (C) 2000-2010 MPlayer Team

and Xorg

[    37.166] 
X.Org X Server 1.8.0
Release Date: 2010-04-02
[    37.186] X Protocol Version 11, Revision 0
[    37.187] Build Operating System: openSUSE SUSE LINUX
[    37.187] Current Operating System: Linux elphaba 2.6.34.4-6-desktop #1 SMP PREEMPT 2010-08-20 19:21:29 +0200 i686
[    37.204] Kernel command line: root=/dev/disk/by-id/scsi-SATA_ST9500325AS_5VE1DQAB-part1 splash=silent resume=/dev/hda2
[    37.204] Build Date: 05 July 2010  09:26:26PM
[    37.204]  
[    37.204] Current version of pixman: 0.18.0

Updating to libdrm, Mesa, xorg-x11-server and xorg-x11-driver-video 
of obs://X11:XOrg might be worth a try. Hope this update doesn't 
make your issues worse.

Two of my computers are also affected by this (really!) annoying bug:
* Samsung Q45 T2390 Laptop
* Asus EEE PC 1000h Netbook
Flash Fullscreen (e.g. YouTube and - which is even worse - southpark.com) freezes after a short while. Video stops, Audio continues a little longer before it freezes, too. The only thing I can do is Power OFF/Power ON my laptop.

Specs:
Samsung Q45:  Intel 965GM
EEE PC 1000h: Intel 945GME
2D driver: intel / 3d driver: Unknown classic 7.8.2
openSUSE 11.3 (i586)
Kernel 2.6.34-12
Flash 10.1.82.76
Firefox 3.6.8

(In reply to comment #68)
> Two of my computers are also affected by this (really!) annoying bug:
> * Samsung Q45 T2390 Laptop
> * Asus EEE PC 1000h Netbook
> Flash Fullscreen (e.g. YouTube and - which is even worse - southpark.com)
> freezes after a short while. Video stops, Audio continues a little longer
> before it freezes, too. The only thing I can do is Power OFF/Power ON my
> laptop.
> 
> Specs:
> Samsung Q45:  Intel 965GM
> EEE PC 1000h: Intel 945GME
> 2D driver: intel / 3d driver: Unknown classic 7.8.2
> openSUSE 11.3 (i586)
> Kernel 2.6.34-12

Upgrade your kernel to 2.6.34.4 (note that the 12 is after a *dash*, not a dot, and don't get me started on that :-), and if that doesn't work, and you're feeling froggy, upgrade your Xorg, as Stefan suggests immediately above (and as I'll do tonight or tomorrow), and see what happens.

Upgrading to 2.6.34.4-0.1 from test repo seems to have fixed it :-)
Thanks

Well, first I would agree with Werner Siedenburg that this is a annoying bug (while being a little bit surprised that this could pass the alpha/beta/rc cycles).
The X11 repo does NOT fix this issue for me, neither did a Kernel update help. Same thing: X11 locks, while the audio goes on playing.

Hardware: 
Lenovo Thinkpad T500 
GPU: Mobile Intel GM45 Express Chipset

Software:
Flash: 10.1.82.76-0.1.1
Firefox: 3.6.8-1.3 (repo: .../mozilla/openSUSE_11.3)

Thanks for your valuable work on this!

Created attachment 386849 [details]
Xorg.0.log saved after a crash.

Ok. Here a detailed description of the freeze:

1. Play flash fullscreen.

2. Picture freezes, sound continues. This can happen within the first few seconds or after five minutes. Very random.

3.0 Now, keyboard still works and there are following possibilities I tried during several freezes:
3.1 ctrl+alt+esc+"click" totally freezes the system/keyboard. Needs hard reboot.
3.2 alt+f4 does nothing.
3.3 esc does nothing.
3.4 ctrl+alt+f1 works fine and brings up the command line. switching back (ctrl+alt+f7) freezes the whole system and needs hard reboot.

Switching to console, one can notice that the Xorg.0.log does not log the freeze (at least in my case).
Attached my Xorg.0.log(.old) from a crash.

(In reply to comment #72)
> Created an attachment (id=386849) [details]
> Xorg.0.log saved after a crash.

There is no crash in this logfile. Likely you need to attach Xorg.0.log.old instead.

Yes, I know. That's why I mentioned it in my post. The attached file is a renamed Xorg.0.log.old. Repeated it several times. Also checked from the command line (point 3.4, post above) with the help of vim the /var/log/Xorg.0.log right after the freeze. I couldn't see more than in the attached file.

Indeed. You don't have a crash, you have a freeze. Still you called it a crash.:-(

Sorry for the confusion ;-) Just checked again:

- Freeze
- Going to command line (ctrl-alt-f1) --> vim /var/log/Xorg.0.log no suspicious entry (like the attached file)
- Returning back to X11 (ctrl-alt-f7): crash/hang or whatever you would call it (keyboard, sound, ... nothing works anymore)
- Hard reset
- Check /var/log/Xorg.0.log.*old*: Still no luck. Just the same as before.

cheers

Forgot something: Without having the flash-fullscreen-freeze, I can safely switch back and forth between command line and X11 without encountering a system hang.

(In reply to comment #73)
> Ok. Here a detailed description of the freeze:
> 
> 1. Play flash fullscreen.
> 
> 2. Picture freezes, sound continues. This can happen within the first few
> seconds or after five minutes. Very random.
> 
> 3.0 Now, keyboard still works and there are following possibilities I tried
> during several freezes:
> 3.1 ctrl+alt+esc+"click" totally freezes the system/keyboard. Needs hard
> reboot.
> 3.2 alt+f4 does nothing.
> 3.3 esc does nothing.
> 3.4 ctrl+alt+f1 works fine and brings up the command line. switching back
> (ctrl+alt+f7) freezes the whole system and needs hard reboot.
> 
> Switching to console, one can notice that the Xorg.0.log does not log the
> freeze (at least in my case).
> Attached my Xorg.0.log(.old) from a crash.


Same EXACTLY here!!

I had the same problem. On top of that, I could always reproduce the freeze. Every time I use google chat, and make the video fullscreen, it would freeze my computer. The only way was Hard reset.

I upgraded the kernel, now I have 2.6.34.4-6 32 bit, and it is working fine now.. No problems so far, It has been around a week.

I'm now quite certain that there are two separate bugs involved here:

1) Flashplayer in fullscreen causes a hard lockup: entire machine falls over, no ping response, if audio was playing, it loops at the hardware buffer size; only a powercycle clears the problem...

which is FIXED by upgrading to the 2.6.34.4 kernel, and is the subject of all three bugs which were marked as dupes of this -- but is *not* the subject of this actual bug, based on the original report, and

2) Flashplayer in fullscreen freezes X; the audio from the video keeps playing continuously, not in a loop, but the video freezes, and (in some cases) the user can switch to a text console--possibly only once--and kill X without having to reboot.  This latter description is pretty clearly the actual bug here on 620157, and if you read all 4 descriptions, you can see that the other three are *not* this particular failing.

Stefan: what's the best solution to this?  Close '157, and open a new bug for the X-only hang?  Or keep this one open, since the kernel upgrade does not reliably fix the X-hang?  (and my apologies, I haven't gotten an X upgrade done on my machine yet to see if doing so fixes the X-hang)

Thanks for the summary, Jay. Adjusted summary to issue 1 now. Could you open a new bug for issue 2, so the still affected persons here can add themselves to Cc
of this new bug? Closing this bug now as fixed.

I rectify that updating the kernel does NOT fix the flash' hang in, BUT its turn out solving the desktop effects' annoying freeze, the describe issue 2 of userr Jay Ashworth has the most looked issues with that I have, butwise thanks for Kernel' advice..

Waiting a fix to flash' hang in!! 

Dell 1545
OpenSUSE 11.3/KDE 4.5

I don't understand why this bug has been marked "resolved".  I have the same Intel graphics chip 965 GM as noted in this report and have upgraded to 2.6.34.7-0.3-desktop kernel with no improvement on fullscreen flash which still image hangs (the picture freezes), locking up the screen on flash fullscreen in firefox and konqueror.  The audio for the flash continues on as normal.  I have also done an Package>All_Packages>Update_if_newer_version_available upgrade in Yast also to no improvement.  Please help with this as in this age of internet tv being limited to a 5" screen in flash regular mode just does not do the business! :)

Rob: See my bug 81.

I have been remiss in opening a new bug for this as Stefan requested; the "video freezes, but audio and rest of computer keeps going" bug is an *app or X level* problem; the "machine locks up hard" bug -- which this bug was opened for -- *is* in fact fixed by a kernel upgrade.

If you have more free time this afternoon than I do, feel free to hang the new bug on "Flash video players freeze X when full screen (but don't lock machine)" noting this bug in that description (and specifically my comment 81), and then xref forwards from here, if you would.  I strongly suspect that that problem is either X or video driver related.

I would do it myself, but I'm in the midst of SATA3G wrangling and my sister wants her DVR back.  :-)

okay,thanks. So this looks like it is a different bug as you suggest so I have opened a new bug here.  https://bugzilla.novell.com/show_bug.cgi?id=641908

Hi,

a week ago I've upgraded to 11.3 and have the same issue. after switching to the full screen mode the X server hangs totally. no signal from video card, no keyboard.

And I have kernel 2.6.34.7-0.5-pae. So I think this bug is still there.. or I missed something?

Thank you,
Roman

Reopening it after the comment.

Kernel 6.36 and 6.37 has fixed my problem, no more crashes!!

Could you please stop reopening this bug again and again? Thanks.

Could you please tell the workaround for supported release 11.3? Thanks.

No, there is none for you. The bug has been fixed for the initial reporters. Not for everyone thinking to see the same issue here.

Dear Stefan,

the problem is I have exactly the same problems are the reporters had. Should I file a new bug then?

Thanks.

(In reply to comment #93)
> the problem is I have exactly the same problems are the reporters had. 

... which I seriously doubt.

> Should I file a new bug then?

Yes.